Struct RawImage

• Defined in File messages.hpp

Struct Documentation

struct RawImage

A raw image.

Public Functions

FoxgloveError encode(uint8_t *ptr, size_t len, size_t *encoded_len)

Encoded the RawImage as protobuf to the provided buffer.

On success, writes the serialized length to *encoded_len. If the provided buffer has insufficient capacity, writes the required capacity to *encoded_len and returns FoxgloveError::BufferTooShort. If the message cannot be encoded, writes the reason to stderr and returns FoxgloveError::EncodeError.

Parameters:

• ptr – the destination buffer. must point to at least len valid bytes.

• len – the length of the destination buffer.

• encoded_len – where the serialized length or required capacity will be written to.

Public Members

std::optional<Timestamp> timestamp

Timestamp of image.

std::string frame_id

Frame of reference for the image. The origin of the frame is the optical center of the camera. +x points to the right in the image, +y points down, and +z points into the plane of the image.

uint32_t width = 0

Image width in pixels.

uint32_t height = 0

Image height in pixels.

std::string encoding

Encoding of the raw image data. See the data field description for supported values.

uint32_t step = 0

Byte length of a single row. This is usually some multiple of width depending on the encoding, but can be greater to incorporate padding.

std::vector<std::byte> data

Raw image data.

For each encoding value, the data field contains image pixel data serialized as follows:

• yuv422 or uyvy:

  • Pixel colors are decomposed into Y’UV channels.

  • Pixel channel values are represented as unsigned 8-bit integers.

  • U and V values are shared between horizontal pairs of pixels. Each pair of output pixels is serialized as [U, Y1, V, Y2].

  • step must be greater than or equal to width * 2.

• yuv422_yuy2 or yuyv:

  • Pixel colors are decomposed into Y’UV channels.

  • Pixel channel values are represented as unsigned 8-bit integers.

  • U and V values are shared between horizontal pairs of pixels. Each pair of output pixels is encoded as [Y1, U, Y2, V].

  • step must be greater than or equal to width * 2.

• nv12:

  • Pixel colors are decomposed into Y’UV channels using 4:2:0 chroma subsampling. The data is stored in NV12 semi-planar layout with two contiguous planes: a Y (luma) plane followed by an interleaved UV (chroma) plane.

  • All channel values are represented as unsigned 8-bit integers.

  • Both planes use step as their row stride.

  • The Y plane contains one luma value per pixel (step * height bytes).

  • The UV plane contains interleaved U, V chroma pairs, subsampled by a factor of 2 in both dimensions (width/2 pairs per row, height/2 rows, step * height/2 bytes). Each U, V pair is shared by a 2x2 block of pixels.

  • width and height must be even.

  • step must be greater than or equal to width.

  • Total data length is step * height * 3/2 bytes.

• rgb8:

  • Pixel colors are decomposed into Red, Green, and Blue channels.

  • Pixel channel values are represented as unsigned 8-bit integers.

  • Each output pixel is serialized as [R, G, B].

  • step must be greater than or equal to width * 3.

• rgba8:

  • Pixel colors are decomposed into Red, Green, Blue, and Alpha channels.

  • Pixel channel values are represented as unsigned 8-bit integers.

  • Each output pixel is serialized as [R, G, B, Alpha].

  • step must be greater than or equal to width * 4.

• bgr8 or 8UC3:

  • Pixel colors are decomposed into Blue, Green, and Red channels.

  • Pixel channel values are represented as unsigned 8-bit integers.

  • Each output pixel is serialized as [B, G, R].

  • step must be greater than or equal to width * 3.

• bgra8:

  • Pixel colors are decomposed into Blue, Green, Red, and Alpha channels.

  • Pixel channel values are represented as unsigned 8-bit integers.

  • Each output pixel is encoded as [B, G, R, Alpha].

  • step must be greater than or equal to width * 4.

• 32FC1:

  • Pixel brightness is represented as a single-channel, 32-bit little-endian IEEE 754 floating-point value, ranging from 0.0 (black) to 1.0 (white).

  • step must be greater than or equal to width * 4.

• bayer_rggb8, bayer_bggr8, bayer_gbrg8, or bayer_grbg8:

  • Pixel colors are decomposed into Red, Blue and Green channels.

  • Pixel channel values are represented as unsigned 8-bit integers, and serialized in a 2x2 bayer filter pattern.

  • The order of the four letters after bayer_ determine the layout, so for bayer_wxyz8 the pattern is:

    w | x
    - + -
    y | z
    

  • step must be greater than or equal to width.

• mono8 or 8UC1:

  • Pixel brightness is represented as unsigned 8-bit integers.

  • step must be greater than or equal to width.

• mono16 or 16UC1:

  • Pixel brightness is represented as 16-bit unsigned little-endian integers. Rendering of these values is controlled in Image panel color mode settings.

  • step must be greater than or equal to width * 2.

Public Static Functions

static Schema schema()

Get the RawImage schema.

The schema data returned is statically allocated.